[zeromq-dev] zmq_poll

Serge Aleynikov serge at aleynikov.org
Sat May 22 01:33:00 CEST 2010

On 5/21/2010 6:52 PM, Chuck Remes wrote:
> On May 21, 2010, at 3:46 PM, Martin Sustrik wrote:
> I'm not really in favor of anything that makes zmq_poll slower. All of the code I am writing with 0mq utilizes non-blocking sockets and polling so I can process things in an asynch fashion.
> My whole goal is to process every socket in a given context within a single thread. If I need more CPU, I spin up a new context in a separate thread and do all of my processing asynchronously. As of now I'm not sharing data between contexts but I clearly could using ipc/inproc transports.
> I hate paying the penalty for memory barriers / mutexes / semaphores / condition variables, etc.

If I understand Martin correctly (and I hope I'm not wrong on this) he's 
not talking about adding any overhead related to introducing mutexes / 
semaphores / condition variables.  It would be helpful to discuss the 
details on this list so that we can have a sense of potential 
implications if there are any.


More information about the zeromq-dev mailing list